PROJECT SUMMARY Medulloblastoma (MB) is the most common malignant primary brain tumor of children. MB development often results from the dysregulation of cellular signaling pathways, such as sonic hedgehog and wingless. Recently, epigenetic aberrations, which represent heritable aberrations in gene expression or cellular phenotypes without changes in DNA sequences, have emerged as a major driving force for tumorigenic events. Histone lysine methylation, a type of histone modification, is a hallmark of epigenetic and transcriptional regulation of gene expression and is modulated by histone methylation modifiers. In contrast to great advances in our understanding of cellular signaling pathways in MB genesis, the pathogenic role of altered histone methylation modifiers in MB development remains largely unknown. Of histone methylations, methylated histone H3 lysine 4 (H3K4) occupies most human gene promoters and is associated with active or poised genes. We previously showed that the H3K4 methyltransferase mixed-lineage leukemia 4 (MLL4; also called MLL2, ALR, and KMT2D) is indispensable for retinoic acid (RA)-induced neuronal differentiation of the model human stem cell line NT2/D1. Consistent with this, we also demonstrated that MLL4 activates the expression of several differentiation-specific genes by depositing methylated H3K4. Our additional results showed that Mll4 brain-specific knockout (BSKO) mice developed spontaneous MBs. These findings are consistent with recent massive sequencing studies of human MBs showing that the MLL4 gene often undergoes somatic mutations and deletions. Our long-term goal is to define the tumor-suppressive role of MLL4 in medulloblastoma pathogenesis. Our analysis of expression data suggests that Mll4-loss-induced MBs are close to the most malignant and metastatic MB subtype Group 3. Based on these definitive findings, our central hypothesis is that MLL4 acts as a tumor-suppressor against MB by activating the expression of tumor suppressor genes via regulation of epigenetic signatures. Here, we propose to study to 1) Assess the role of MLL4 in MB development using genetically engineered mouse models; 2) Determine the molecular mechanism underlying the genesis of Mll4-loss-driven MB; 3) Characterize the effect of Mll4 loss on epigenetic signatures during MB genesis. These studies will reveal the previously unknown epigenetic mechanism underlying MB pathogenesis and provide beneficial information for the development of MB therapies.